Radiotherapy is an important component of multi-modality treatment strategies for childhood cancers. However, radiotherapy it is known to cause a diverse spectrum of side-effects including second cancers, which can be physically and emotionally devastating. Our long term goal is to reduce the incidence of second cancers by developing a user-friendly model that will enable clinicians to select the type of radiotherapy with the lowest risk of second cancer. It is our hypothesis that using proton therapy instead of photon therapy will significantly reduce the predicted risk of second cancers that develop in the breast and thyroid of children treated for Hodgkin's disease and medulloblastoma, respectively. Testing this hypothesis requires knowledge of the radiation exposures to organs at risk from contemporary, multi-modality treatments. An innovative contribution of this project is the development of models to predict stray radiation exposures incurred during proton therapy, a problem that is not well understood. The following Specific Aims will test the hypothesis: 1) Identify the dosimetric factors that govern the stray radiation exposures to individual patients. We will use the method of Monte Carlo simulation to determine how strongly each parameter influences the stray radiation exposures. 2) Establish the confidence interval that can be achieved with an integrated dose/risk model for predicting second cancer development for individual patients. The methods to validate organ doses from analytical-model predictions will include comparison with Monte Carlo simulations and measurements. We will also carry out a comprehensive uncertainty analysis of the risk predictions. 3) In virtual clinical trials comparing photon therapy versus proton therapy, establish the integrated dose/risk model's ability to detect a significant difference in the predicted incidence of second cancer. To achieve this aim, we will perform virtual clinical trials (for the treatment of medulloblastoma and Hodgkin's disease) that compare the risk of developing second cancers of the thyroid and breast after proton therapy versus photon therapy. Public Health Relevance: Today approximately 80% of children survive their initial cancer but are at elevated risk for developing second cancers later in life. The proposed research will lead to individualized treatment strategies that will reduce second-cancer risks. In the future, we expect that the lower-risk treatments will result in fewer second cancers and this will translate into longer, higher-quality lives for survivors of childhood cancer (presently 270,000 persons in the United States).